The Icon Bar: Castle Technology deny GPL breach

Castle Technology deny GPL breach

The Icon Bar has just recieved a press release from Mike Williams, on behalf of Castle Technology in response to the recent claims that they had used code which was protected by the General Public Licence.

In the Press Release, they state the following:

"The RISC OS 5.00 kernel did not contain work taken from or derived from the ARM-Linux or Linux kernel."
"The RISC OS 5.01 kernel did not contain work taken from or derived from the ARM-Linux or Linux kernel."
"The RISC OS 5.02 kernel does not contain work taken from or derived from the ARM-Linux or Linux kernel."
"There are no plans to use GPL derived code in any part of the RISC OS kernel in the future."

The press release goes on to state that "For the avoidance of doubt, the hardware abstraction layer (roughly analogous to a PC's BIOS) has it's PCI allocation and bridge setup based in part on the following functions from the Linux kernel sources:"

pci_alloc_primary_bus
pbus_size_bridges
pbus_assign_resources_sorted
pci_setup_bridge
pci_bridge_check_ranges
pbus_size_mem
pbus_assign_resources
pci_assign_unassigned_resources
pci_scan_bus
pcibios_update_resource
pci_read_bases
pci_alloc_bus
pci_add_new_bus
pci_do_scan_bus
pci_scan_bridge
pci_setup_device
pci_scan_device
pci_scan_slot
pcibios_fixup_bus
pci_calc_resource_flags
pci_size
pdev_fixup_device_resources
pbus_assign_bus_resources
pci_do_scan_bus
pcibios_fixup_pbus_ranges
pci_assign_resource
pdev_sort_resources
pdev_enable_device
pbus_size_io

Castle state that "any company or individual wishing to recieve a copy of the source code to this component should apply in writing to:"

The Managing Director
Castle Technology Ltd
Ore Trading Estate
Woodbridge Road
Framlingham
Suffolk
IP13 9LL

You will also need to enclose a formatted 3.5" floppy diskette and return postage stamps (or international reply coupons if you are outside the UK)

"These sources will also form an integral part of a forthcoming Linux port to the IYONIX" they go on to state.

Finally, they close by saying "With the tough goal of fitting all of the supporting software and applications for the IYONIX computer into just 4Mbytes of ROM, later issues of the supporting software have had to have function names removed (along with a strategy of tokenising textual messages and compressing binaries) to make room for, in particular, the support for the 'boot keyboard' USB drivers."

Update: for further debate on this subject, head over to Slashdot.

Castle Technology deny GPL breach

This is a long thread. Click here to view the threaded list.

Paul Stewart

Message #91729, posted at 19:23, 10/2/2003

Unregistered user

At least that settles it!!!!
No wrong doing on behalf of Castle
All that's needed now is an apology from the Linux people.

[ Log in to reply ]

Moose

Message #91730, posted at 19:25, 10/2/2003, in reply to message #91729

Unregistered user

WHAT settles it? CTL: "We did no wrong, guv" Judge: "Ok, if you say so, you're free to go" -- what a load of crap!

[ Log in to reply ]

GuestX

Message #91731, posted by guestx at 19:28, 10/2/2003, in reply to message #91730

Member
Posts: 102

More investigation required by Dr Blix, erm Justin Fletcher, and his team, I think. On the one hand you have a declaration of no Linux kernel code being used, but on the other, certain things are "based in part" on Linux kernel sources.

Now, if it had been said that the function *names* used in RISC OS 5.x had been based on those employed in the Linux kernel sources, then Castle would be pretty instantly off the hook (excluding the dreadful possibility that they are not giving a full declaration of their weapons programmes, erm source code origins). But that isn't what has been said.

To top it all off, we get offered the source code - something which wouldn't have happened if this GPL furore hadn't come up. Add to that the "integral part of a forthcoming Linux port" and I'm sure it wouldn't require an invitation to be issued to Donald Rumsfeld to find someone who isn't entirely convinced by this particular development.

[ Log in to reply ]

Anonymous

Message #91732, posted at 19:29, 10/2/2003, in reply to message #91731

Unregistered user

Personally I'll wait and see what Justin Fletcher has
to say in response. He is a well respected member of
the RISC OS community and his words also carry much
weight.

[ Log in to reply ]

Peter Naulls

Message #91733, posted by pnaulls at 19:34, 10/2/2003, in reply to message #91732

Member
Posts: 317

Yay, verbatim press release. Well done TIB :-)

[ Log in to reply ]

Andrew Poole

Message #91734, posted by andypoole at 19:35, 10/2/2003, in reply to message #91733

Mouse enthusiast
Web
Twitter

Posts: 5558

Yay! pointless comment! well done Peter

[ Log in to reply ]

John Hoare

Message #91735, posted by moss at 19:35, 10/2/2003, in reply to message #91734

...óflå...óflå...óflå...

Posts: 9348

/me slaps you both :-p

[ Log in to reply ]

SimonC

Message #91736, posted at 19:36, 10/2/2003, in reply to message #91735

Unregistered user

All that says is there's nothing in the RO kernel, which suggests that the PCI code is not part of the kernel, but somewhere else.

To be honest I was hoping all along that this was a big mistake, but that statement makes it look less likely. Just what does "based on" mean?

[ Log in to reply ]

Andrew Sidwell

Message #91737, posted by takkaria at 19:36, 10/2/2003, in reply to message #91736

Member
Posts: 324

One presumes CTL mean that they've based their code interface on the Linux header files.

As the header files are an interface, they are not breaching the GPL for using interfaces is allowed.

BTW: does the Iyonix still have Paint, Edit and Draw in ROM? If so, they might want to make them disc-based instead and have them on the HD.

[ Log in to reply ]

Gavin Smith

Message #91738, posted by SparkY at 19:36, 10/2/2003, in reply to message #91737

Danger! Danger! High Voltage!
Posts: 697

It's what I have been waiting to read all day :) I also look forward to Drobe's take on it cos, frankly, I don't fully understand it. I don't think they did a great job in their "avoidance of doubt"...

[ Log in to reply ]

Liyonix User

Message #91739, posted at 19:42, 10/2/2003, in reply to message #91738

Unregistered user

Maybe RISC OS 5 is all fake. Perhaps it's actually just Linux running the fake RISC OS front end?!

[ Log in to reply ]

John Hoare

Message #91740, posted by moss at 19:43, 10/2/2003, in reply to message #91739

...óflå...óflå...óflå...

Posts: 9348

Yes; I was confused about it as well. I hope takka is correct.

[ Log in to reply ]

GuestX

Message #91741, posted by guestx at 19:45, 10/2/2003, in reply to message #91740

Member
Posts: 102

Well, Liyonix User - hmmm, is Liyonix a bit like Lindows? ;-) - if they'd done that, I can imagine that a fair number of people would be rather interested. Even I and a number of cynical cohorts would be tempted to give it a spin, not to mention all those people on Slashdot...

[ Log in to reply ]

Andrew Sidwell

Message #91742, posted by takkaria at 19:48, 10/2/2003, in reply to message #91741

Member
Posts: 324

I was kind of hoping that people would recognize the trolling attempt by the "Liyonix user" and ignore him. Sadly not.

Be sensible, if anyone says RO in any form is Linux with a RO fake front-end, ignore them. They're trolls, they live off pointless replies.

[ Log in to reply ]

Jeffrey Lee

Message #91743, posted by Phlamethrower at 19:51, 10/2/2003, in reply to message #91742

Hot Hot Hot Hot Hot Hot Hot Hot Hot Hot Hot Hot Hot stuff

Posts: 15100

"Interesting."

/me munches some popcorn and waits for the next round...

[ Log in to reply ]

Annraoi

Message #91744, posted by ams at 19:54, 10/2/2003, in reply to message #91743

Member
Posts: 56

The above seems a satisfactory enough explanation. As the source for the PCI/HAL will be available that should allow an A/B comparison of that source with the sources from Linux (whatever version Justin used) to be made.

This hopefully finally sorts this matter out.

Regards

Annraoi

[ Log in to reply ]

Stinky Wizzleteats

Message #91745, posted at 19:55, 10/2/2003, in reply to message #91744

Unregistered user

"Loosely based on", in that substantial chunks compile the same. How odd.

3.5" disc? What's this? The stone age? Or do they need some time to reformat the source... and if it's *nothing* to do with GPL why are they releasing it to anyone?

Something steeenks.

[ Log in to reply ]

Simon Wilson

Message #91746, posted by ksattic at 19:58, 10/2/2003, in reply to message #91745

Finally, an avatar!

Posts: 1291

LOL, you could probably fit most of the source for RISC OS onto a single 3.5" disk.

I think we should reserve further judgement until someone gets a chance to compare the source from Castle with the source from Linux.

[ Log in to reply ]

Stinky Wizzleteats

Message #91747, posted at 19:59, 10/2/2003, in reply to message #91746

Unregistered user

Equally the "The RISC OS 5.00 kernel did not contain work taken from or derived from the ARM-Linux or Linux kernel." lines are dodgy. The kernel could be defined in many ways, and the questioned code could be easily claimed to be external from the kernel.

[ Log in to reply ]

Darren Winsper

Message #91748, posted at 20:03, 10/2/2003, in reply to message #91747

Unregistered user

I still think they're in violation. If the code is based in part on GPL code, then the distributed product itself falls under the GPL. See Section 2b in particular:

b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

Castle have admitted their product in part contains code derived from Linux, which is under the GPL. Seems pretty clear-cut to me.

[ Log in to reply ]

Phlip

Message #91749, posted at 20:04, 10/2/2003, in reply to message #91748

Unregistered user

Moo

[ Log in to reply ]

Andrew Sidwell

Message #91750, posted by takkaria at 20:04, 10/2/2003, in reply to message #91749

Member
Posts: 324

Darren: the header files are merely an interface; using an interface is not a breach of the GPL.

See above on this news item.

[ Log in to reply ]

GuestX

Message #91751, posted by guestx at 20:05, 10/2/2003, in reply to message #91750

Member
Posts: 102

Good point, SW. In /usr/src/linux-xxx, the PCI stuff doesn't *all* appear in the kernel directory - some of it is in the drivers directory as "pci", alongside the "acorn" directory, although I hope Castle doesn't need to borrow Acorn Podule support from Linux. :-)

[ Log in to reply ]

Simon Wilson

Message #91752, posted by ksattic at 20:07, 10/2/2003, in reply to message #91751

Finally, an avatar!

Posts: 1291

Castle have admitted their product in part contains code derived from Linux

No they haven't - where did you read that?

[ Log in to reply ]

Darren Winsper

Message #91753, posted at 20:08, 10/2/2003, in reply to message #91752

Unregistered user

I quote:

the hardware abstraction layer (roughly analogous to a PC's BIOS) has it's PCI allocation and bridge setup based in part on the following functions from the Linux kernel sources:

Where does it say they only used headers?

[ Log in to reply ]

Andrew Sidwell

Message #91754, posted by takkaria at 20:09, 10/2/2003, in reply to message #91753

Member
Posts: 324

Darren: look, they say they haven't used GPL'd code. Therefore we assume that they've used headers only, which means that the information they've given us makes sense, and if we believe them, then they're not breaching the GPL.

[ Log in to reply ]

Simon Wilson

Message #91755, posted by ksattic at 20:12, 10/2/2003, in reply to message #91754

Finally, an avatar!

Posts: 1291

We must give Castle the benefit of the doubt till someone compares the sources.

It's not breaking copyright to base your work upon someone elses, unless the product/process in question is patented. It is against the law to directly derive your work from someone elses if the work is copyrighted. However, Castle have denied this and we must wait to examine their proof.

[ Log in to reply ]

Stinky Wizzleteats

Message #91756, posted at 20:12, 10/2/2003, in reply to message #91755

Unregistered user

Merely copying function headers would not result in identical compiled output. You'd need to duplicate structures, and, oops, code.

They're up shite creek without a paddle, engine etc, and if their boat has as many holes as that press release, they're shafted.

[ Log in to reply ]

Darren Winsper

Message #91757, posted at 20:17, 10/2/2003, in reply to message #91756

Unregistered user

Bloody hell, it's right there in the press release. They say the kernel doesn't have any GPL code in it, but the hardware abstraction layer has code based off Linux PCI code. Why would they say it's based off Linux code if it just had a (largely) compatible interface?!

[ Log in to reply ]

Andrew Sidwell

Message #91758, posted by takkaria at 20:19, 10/2/2003, in reply to message #91757

Member
Posts: 324

Darren: because that's what "based on" can mean, funnily enough. I've seen it used that way before, and it's perfectly normal.

[ Log in to reply ]

Pages (3): 1 > >|